Work

Accurate projector calibration based on an improved sub-pixel homography mapping method of circular edge points

Year: 2024

Type: article

Abstract: Fringe projection profilometry (FPP) is an efficient and accurate three-dimensional measurement technology. The premise of measurement is to calibrate the parameters of the camera and projector setup in the FPP system. However, the projector cannot capture images. Thus, phase mapping between the camera and projector is employed to provide the coordinates of the feature points for projector calibration. The calibration accuracy is determined by the precision of the mapping. In this paper, an improved sub-pixel homography mapping method of circular edge points is proposed. The phase errors of the circular edge points are suppressed by homography constraint. The accurate sub-pixel mapping relationship between the camera pixels and the projector pixels is established based on homography estimation. Calibration experiment was designed to validate the feasibility of proposed method. Furthermore, the calibration results were verified by measurement experiments of out-of-plane displacement, in-plane distance, diameter and distance of two standard ceramic spheres. In addition, the number of points for homography estimation is not manually specified, which is different from the methods that adopt neighborhood windows for error correction or smoothing. Compared to the existing calibration methods, the proposed method offers advantages in terms of accuracy and efficient for projector calibration.

Source: Optics and Lasers in Engineering

DOI: 10.1016/j.optlaseng.2024.108768

Authors: Sen Wang, Bin Liu, Lin Li, Haoxuan Zhu, Ruoyu He, Guanhao Wu

Institutions: Tianjin University of Technology, Institute of Spacecraft System Engineering, Tsinghua University

Language: English

Topic: Optical measurement and interference techniques

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,860